Urinary tract infections and asymptomatic bacteriuria in pregnancy

INTRODUCTION — Urinary tract infections (UTIs) are common in pregnant women. By convention, UTI is defined either as a lower tract (acute cystitis) or upper tract (acute pyelonephritis) infection. UTIs (acute cystitis and pyelonephritis) and asymptomatic bacteriuria in pregnant women will be reviewed here.

Issues related to UTIs or asymptomatic bacteriuria in other populations are discussed in detail elsewhere. (See "Acute simple cystitis in adult and adolescent females" and "Acute simple cystitis in adult and adolescent males" and "Acute complicated urinary tract infection (including pyelonephritis) in adults and adolescents" and "Asymptomatic bacteriuria in adults" and "Catheter-associated urinary tract infection in adults".)

EPIDEMIOLOGY

Incidence and risk factors — The incidence of bacteriuria in pregnant women is approximately the same as that in nonpregnant women; however, recurrent bacteriuria is more common during pregnancy. Additionally, the incidence of pyelonephritis is higher than in the general population, likely as a result of physiologic changes in the urinary tract during pregnancy. (See 'Pathogenesis' below.)

Asymptomatic bacteriuria occurs in 2 to 7 percent of pregnant women [1,2]. It typically occurs during early pregnancy, with only approximately a quarter of cases identified in the second and third trimesters [3]. Factors that have been associated with a higher risk of bacteriuria include a history of prior urinary tract infection, pre-existing diabetes mellitus, and low socioeconomic status [4,5].

Without treatment, as many as 20 to 35 percent of pregnant women with asymptomatic bacteriuria will develop a symptomatic urinary tract infection (UTI), including pyelonephritis, during pregnancy [6,7]. This risk is reduced by 70 to 80 percent if bacteriuria is eradicated (see 'Rationale for treatment' below). Although a study from the Netherlands suggested a low rate of pyelonephritis among 208 women with untreated asymptomatic bacteriuria (2.4 percent versus 0.6 percent among 4035 women without bacteriuria), this study included only low-risk women with uncomplicated singleton pregnancies without diabetes mellitus or urinary tract abnormalities, and it is uncertain whether these results are generalizable [8].

Acute cystitis occurs in approximately 1 to 2 percent of pregnant women, and the estimated incidence of acute pyelonephritis during pregnancy is 0.5 to 2 percent [9-12]. Most cases of pyelonephritis occur during the second and third trimesters. As an example, the incidence of acute pyelonephritis in pregnancy in the setting of routine prenatal screening for asymptomatic bacteriuria was examined in a prospective study of a general obstetric population [11]. During the two year study period, 440 cases of acute pyelonephritis were identified in 32,282 pregnant women (14 per 1000 deliveries). The majority of cases occurred in the second trimester (53 percent). In addition to prior untreated bacteriuria, other clinical characteristics that have been associated with acute pyelonephritis during pregnancy include age <20 years, nulliparity, smoking, late presentation to care, sickle cell trait, and pre-existing (not gestational) diabetes [11-13]

Pregnancy outcomes — Many studies have described a correlation between maternal urinary tract infection, particularly asymptomatic bacteriuria, and adverse pregnancy outcomes. Studies have also suggested that acute pyelonephritis has a similar association, but there are several variables that potentially confound this association, such as socioeconomic status and previous preterm delivery.

Untreated bacteriuria has been associated with an increased risk of preterm birth, low birth weight, and perinatal mortality in most [1,6,14-19], but not all [8,20], studies. As an example, in a meta-analysis of 19 studies, among women without bacteriuria, the risks of preterm birth and a low birth weight infant were one-half and two-thirds the risks among women with asymptomatic bacteriuria [19]. Other pregnancy complications have also been associated with bacteriuria. As an example, a case control study of over 15,000 pregnant women found an increased risk of preeclampsia with either asymptomatic bacteriuria or symptomatic UTI [21].

No correlation has been clearly established between acute cystitis of pregnancy and increased risk of low birth weight, preterm delivery, or pyelonephritis [16], perhaps because pregnant women with symptomatic lower UTI usually receive treatment.

Pyelonephritis, however, has been associated with adverse pregnancy outcomes. In an 18-year retrospective study of over 500,000 singleton pregnancies followed at a large health care system in the United States, the rate of preterm birth, primarily between weeks 33 and 36, was higher among the 2894 women who had pyelonephritis during pregnancy (10.3 versus 7.9 percent among those who did not, OR 1.3, 95% CI 1.2-1.5) [12]. There were no differences in stillbirth or neonatal death. Other complications of pyelonephritis include anemia, sepsis, and respiratory distress [11]. Maternal morbidity and obstetric outcomes with pyelonephritis do not appear to differ by trimester [22].

PATHOGENESIS — The organisms that cause bacteriuria and urinary tract infections (UTI) in pregnant women are of the same species and have similar virulence factors as in nonpregnant women. Thus, the basic mechanism of entry of bacteria into the urinary tract is likely to be the same for both groups [23]. However, the smooth muscle relaxation and subsequent ureteral dilatation that accompany pregnancy are thought to facilitate the ascent of bacteria from the bladder to the kidney, resulting in the greater propensity for bacteriuria to progress to pyelonephritis during pregnancy [14,24]. (See "Maternal adaptations to pregnancy: Renal and urinary tract physiology".)

Pressure on the bladder and ureters from the enlarging uterus may also increase the risk of progression to pyelonephritis. In addition, the immunosuppression of pregnancy may contribute. As an example, mucosal interleukin-6 levels and serum antibody responses to Escherichia coli antigens appear to be lower in pregnant women [25].

MICROBIOLOGY — As in nonpregnant women, E. coli is the predominant uropathogen found in both asymptomatic bacteriuria and urinary tract infection (UTI) in pregnant women. As an example, in a study of over 400 cases of pyelonephritis, E. coli accounted for approximately 70 percent of cases [11]. Other organisms responsible for infection included Klebsiella and Enterobacter species (3 percent each), Proteus (2 percent), and gram-positive organisms, including group B Streptococcus (10 percent). Group B Streptococcus in pregnancy is discussed in detail elsewhere. (See "Group B streptococcal infection in pregnant individuals", section on 'Urinary tract'.)

As in other community-acquired infections, antimicrobial resistance is an increasing concern. Infections caused by extended-spectrum beta-lactamase (ESBL)-producing strains are increasing in number [26,27]. In India, ESBL-producing uropathogens are a particular problem, even in pregnant women [28].

Isolation of more than one species or the presence of Lactobacillus or Cutibacterium (formerly Propionibacterium) acnes may indicate a specimen contaminated by vaginal or skin flora. However, repeated isolation of Lactobacillus with high colony counts and without other organisms in consecutive urine cultures may not represent simple contamination, although the significance of this finding in pregnancy is not known.

ASYMPTOMATIC BACTERIURIA

Diagnosis — The diagnosis of asymptomatic bacteriuria is made by finding high-level bacterial growth on urine culture in the absence of symptoms consistent with urinary tract infection (UTI). Details on the timing of screening, specimen collection, and diagnostic criteria are discussed below.

Screening — We agree with the guidelines from the Infectious Diseases Society of America that recommend screening all pregnant women for asymptomatic bacteriuria at least once in early pregnancy [2]. Other expert groups make a similar recommendation [29,30]. The rationale for screening is the same as for treatment of bacteriuria and is discussed elsewhere. (See 'Rationale for treatment' below.)

Screening for asymptomatic bacteriuria is performed at the first prenatal visit with a urine culture [2,31]. Rescreening among those who did not have bacteriuria on the initial test is generally not performed in low-risk women. It is reasonable to rescreen women at high risk for infection (eg, history of UTI or presence of urinary tract anomalies, diabetes mellitus, hemoglobin S, or preterm labor), however the optimal target populations for this is uncertain. There is minimal evidence informing the benefits and harms of repeat screening following an initial negative culture.(See "Prenatal care: Initial assessment".)

Specimen collection — The diagnosis of asymptomatic bacteriuria should be based on culture of a urine specimen collected in a manner that minimizes contamination. Although the optimal method for collecting voided urine is uncertain, instructing women to spread their labia and collect a midstream urine (the second portion of the voided urine after discarding the initial stream) without requiring a clean-catch (local cleansing of the urethral meatus and surrounding mucosa) seems most reasonable. Routine catheterization to screen for bacteriuria is not warranted due to the risk of introducing infection. (See "Sampling and evaluation of voided urine in the diagnosis of urinary tract infection in adults".)

Several studies suggest that local cleansing is of little value, as it does not reduce contamination when a midstream urine is collected [32-34]. As an example, in a study of 113 asymptomatic pregnant women, each was instructed to collect a sample from the first concentrated morning urine, a midstream sample, and a clean-catch midstream sample, in that order, over the course of a day [32]. Rates of mixed growth and growth of skin flora on culture in midstream urine were comparable with those observed in the morning and clean-catch samples. Overall rates of contamination were high in all three samples, and the women were tested at a mean of 32 weeks of gestation as opposed to the recommended period of 16 weeks.

Proper handling and processing of the specimen is crucial to avoid false-positive results. (See "Microbiology specimen collection and transport".)

Diagnostic criteria — For asymptomatic women, bacteriuria is formally defined as two consecutive voided urine specimens with isolation of the same bacterial strain in quantitative counts of ≥10⁵ colony-forming units (cfu)/mL or a single catheterized urine specimen with one bacterial species isolated in a quantitative count of ≥10² cfu/mL [2]. In clinical practice, however, only one voided urine specimen is typically obtained, and diagnosis (and treatment initiation) is made in women with ≥10⁵ cfu/mL without obtaining a confirmatory repeat culture. The diagnosis (and treatment) of asymptomatic bacteriuria due to group B streptococcus during pregnancy is discussed in detail elsewhere. (See "Group B streptococcal infection in pregnant individuals", section on 'Asymptomatic bacteriuria'.)

If bacteria that are not typical uropathogens (such as lactobacillus) are isolated, treatment should be reserved for patients in whom the organism grows as a single isolate on consecutive cultures.

Rapid screening tests, such as dipstick, enzymatic screen, reagent strip, or interleukin-8 testing, do not come close to urine culture in terms of sensitivity and specificity for detecting asymptomatic bacteriuria in pregnant women and should not be used [35-37]. In addition, cultures are useful in guiding therapy. This can be particularly important in pregnancy, during which the number of safe treatment alternatives is reduced.

Management

Rationale for treatment — Asymptomatic bacteriuria during pregnancy increases the risk of pyelonephritis and has been associated with adverse pregnancy outcomes, such as preterm birth and low birth weight infants (see 'Epidemiology' above). Antimicrobial treatment reduces the risk of subsequent development of pyelonephritis and is associated with improved pregnancy outcomes [6,38-43]. This was illustrated in a meta-analysis of 15 randomized trials of antibiotic treatment versus placebo or no treatment for pregnant women with asymptomatic bacteriuria [6]. Antibiotic therapy was more likely to clear asymptomatic bacteriuria (odds ratio [OR] 0.30, 95% CI 0.18-0.53) and to lower the incidence of pyelonephritis (OR 0.24, 95% CI 0.13-0.41). There was also a reduction in the incidence of low birth weight infants with antibiotic treatment. However, the included studies that evaluated these outcomes were deemed to be of poor quality.

The meta-analysis described above included a trial of pregnant women in the Netherlands (where screening for asymptomatic bacteriuria is not performed routinely), which suggested generally similar effects of treatment [8]. Women with asymptomatic bacteriuria between 16 and 22 weeks gestation who were not treated or received placebo treatment had a higher rate of pyelonephritis (2.4 percent of 208 women) compared with both those who received nitrofurantoin treatment for asymptomatic bacteriuria (0 percent of 40 women) and those who did not have asymptomatic bacteriuria (0.6 percent of 4035 women). Likewise, low birth weight occurred in 17 of 208 untreated or placebo-treated women with asymptomatic bacteriuria (8 percent) compared with 1 of 40 of nitrofurantoin-treated women (2.5 percent). While this difference was not statistically significant, the study was under-powered for this outcome. Preterm birth did not differ between these two groups. Notably, the study included only women who were at low risk for UTI, pre-term birth, or other complications, and the pyelonephritis rate was lower than in previous studies of pregnant women with asymptomatic bacteriuria [20].

Antimicrobial treatment — Asymptomatic bacteriuria is treated with an antibiotic tailored to the susceptibility pattern of the isolated organism, which is generally available at the time of diagnosis. Potential options include beta-lactams, nitrofurantoin, and fosfomycin (table 1). Although nitrofurantoin is generally avoided during the first trimester, it is an appropriate alternative when other options cannot be used. Safety of antimicrobial agents used for bacteriuria during pregnancy is discussed elsewhere. (See 'Antibiotic safety in pregnancy' below.)

The optimal duration of antibiotics for asymptomatic bacteruria is uncertain. We typically give five to seven days of therapy. Short courses of antibiotics are preferred to minimize the antimicrobial exposure to the fetus. Short course antibiotic therapy is usually effective in eradicating asymptomatic bacteriuria of pregnancy, although single-dose regimens may not be as effective as slightly longer regimens [44-46]. As an example, in a meta-analysis of 13 studies comparing single-dose treatment with four to seven day treatments, there was a trend towards lower rates of bacteriuria clearance with the single-dose regimen [46].

An exception is single-dose fosfomycin, which successfully treats bacteriuria. In three trials comparing this drug with other therapies administered for a longer time, eradication of the organism was comparable (77 to 94 percent versus 68 to 94 percent with other agents) [47].

Uncertain role of follow-up testing — Up to 30 percent of women fail to clear asymptomatic bacteriuria following a short course of therapy [1]. However, given the paucity of data, it is not clear that there is any clinical benefit to performing a follow-up culture after an initial episode of asymptomatic bacteriuria or retreating recurrent asymptomatic bacteriuria. Thus, we do not recommend repeat urine culture following treatment for asymptomatic bacteriuria or repeat treatment if asymptomatic bacteriuria is detected on repeat screening.

Statements from expert groups also highlight the uncertain value of repeat screening for asymptomatic bacteriuria. The 2019 Infectious Diseases Society of America (IDSA) guidelines reported that there is no direct evidence evaluating the benefit of repeated screening following an initial episode of asymptomatic bacteriuria and that it is not known whether retreatment of recurrent or persistent bacteriuria improves outcomes [2]. They also note insufficient evidence evaluating the benefits or risks of prophylactic antimicrobial therapy for the duration of the pregnancy to prevent recurrent asymptomatic bacteriuria. The American College of Obstetricians and Gynecologists (ACOG) guidelines state that follow-up urine culture after treatment of asymptomatic bacteriuria is recommended in the literature but indicate that more data are needed to determine the effectiveness of this strategy [48].

ACUTE CYSTITIS

Clinical manifestations — Cystitis is a symptomatic infection of the bladder. The typical symptoms of acute cystitis in the pregnant woman are the same as in nonpregnant women and include the sudden onset of dysuria and urinary urgency and frequency. Hematuria and pyuria are also frequently seen on urinalysis. (See "Acute simple cystitis in adult and adolescent females", section on 'Clinical manifestations'.)

Systemic symptoms, such as fevers and chills, are absent in acute cystitis (table 2).

Diagnosis — Acute cystitis should be suspected in pregnant women who complain about dysuria. Although urinary frequency and urgency are typical findings of acute cystitis, they are also frequently a normal physiologic change of pregnancy and reported by pregnant women without cystitis or bacteriuria [49,50]. The presence of fever and chills, flank pain, or costovertebral angle tenderness should raise suspicion for pyelonephritis (see 'Acute pyelonephritis' below). A urinalysis and urine culture should be performed in pregnant women who have new onset dysuria. Specimen collection is the same as for asymptomatic bacteriuria. (See 'Specimen collection' above.)

The diagnosis of acute cystitis is confirmed by finding of bacterial growth on urine culture. Prior to confirming the diagnosis, empiric treatment is typically initiated in a patient with consistent symptoms and pyuria on urinalysis (see 'Management' below). As in nonpregnant women, pyuria is usually present in almost all pregnant women with symptomatic urinary tract infection, and its absence strongly suggests an alternative diagnosis. (See "Sampling and evaluation of voided urine in the diagnosis of urinary tract infection in adults", section on 'Interpretation of pyuria'.)

Studies defining the threshold of bacterial growth on voided urine that represents significant bacteriuria in pregnant women have not been performed, but based on studies in nonpregnant women, relatively low colony counts can reflect true bacteriuria in the presence of symptoms. In nonpregnant women with acute simple cystitis, coliform colony counts in voided urine as low as 10² colony-forming units (cfu)/mL have been noted to reflect bladder infection [51-53]. As most clinical laboratories do not routinely quantify urine isolates to 10² cfu/mL, it is reasonable to use a quantitative count ≥10³ cfu/mL in a symptomatic pregnant woman as an indicator of symptomatic UTI. If bacteria that are not typical uropathogens (such as lactobacillus) are isolated, the diagnosis of cystitis is typically made only if they are isolated in high bacterial counts (≥10⁵ cfu/mL). (See "Sampling and evaluation of voided urine in the diagnosis of urinary tract infection in adults", section on 'Definition of a positive culture'.)

Differential diagnosis — As in nonpregnant women, dysuria in pregnant women can be a result of other infectious and noninfectious processes, such as vaginitis or urethritis. Similarly, urinary frequency and urgency may be symptoms of normal pregnancy in the absence of urinary tract infection. However, true bacteriuria is typically not present in these settings and thus distinguishes acute cystitis. If not already performed, testing for sexually transmitted infections (such as chlamydia and gonorrhea) is warranted for pregnant women with dysuria without bacteriuria or women who have persistent dysuria despite successful treatment of bacteriuria. (See "Acute simple cystitis in adult and adolescent females", section on 'Differential diagnosis'.)

Management

Antimicrobial treatment — Antibiotic treatment of acute cystitis in pregnant women is often empiric, initiated at the time of complaints of dysuria, and then tailored to the susceptibility pattern of the isolated organism once urine cultures return.

●Empiric regimen selection – For empiric therapy, we typically choose between cefpodoxime, amoxicillin-clavulanate, and fosfomycin, given their safety in pregnancy and the somewhat broader spectrum of activity compared with other agents (such as amoxicillin or cephalexin) (table 1). Nitrofurantoin and trimethoprim-sulfamethoxazole are typically avoided during the first trimester (and also near term for trimethoprim-sulfamethoxazole); however, overall evidence suggests that they are appropriate alternatives, even during these periods, if other options cannot be used (eg, because of drug allergy). This is consistent with a statement from ACOG [54]. The choice between the options should be individualized on the basis of several factors, including prior microbiological data, patient allergy history, local practice patterns, local community resistance prevalence, availability, and cost [55]. Safety of antimicrobial agents used for cystitis during pregnancy are discussed elsewhere. (See 'Antibiotic safety in pregnancy' below.)

For women who are thought to be at risk for or have documented infection with extended-spectrum beta-lactamase (ESBL)-producing Enterobacteriaceae, nitrofurantoin and fosfomycin are active in vitro against many such strains and are potential oral options [27,56].

Although there are limited data in pregnant women, a meta-analysis suggested that there are no large differences in outcomes between different antibiotic regimens in terms of cure rates, recurrent infection, incidence of preterm delivery, and the need for a change of antibiotics [57]. All of the antibiotics studied were very effective and complications were rare. There was not enough evidence to recommend a particular treatment scheme.

●Duration – The optimal duration of treatment for acute cystitis during pregnancy is uncertain. As with asymptomatic bacteriuria, short courses of antibiotics are preferred, to minimize the antimicrobial exposure to the fetus. Our approach depends on the agent used and ranges from a single dose of fosfomycin to a three- to seven-day course for other antibiotics (table 1), as long as there are no symptoms suggestive of pyelonephritis (eg, flank pain, nausea/vomiting, fever [>38°C], and/or costovertebral angle tenderness). Based on data among nonpregnant individuals, there appear to be no differences between short antibiotic courses (three to seven days) and longer ones [1,55,58].

Follow-up — As long as symptoms have resolved with treatment, we do not routinely check a follow-up culture. As above, the benefits of repeated screening and treatment of asymptomatic bacteriuria are uncertain. (See 'Uncertain role of follow-up testing' above.)

If the initial urine culture isolate is resistant to the empirically selected agent and the patient continues to have cystitis symptoms, we adjust the antibiotic regimen to target the pathogen. For those whose symptoms have resolved, we do not routinely adjust the empiric antibiotic regimen. However, there are no direct data informing the risks or benefits of this approach, and other clinicians may reasonably opt to adjust the regimen in this setting.

Management of recurrent cystitis — The optimal approach to recurrent cystitis during pregnancy is uncertain. For individuals who have three or more episodes of recurrent cystitis during pregnancy, we favor antimicrobial prophylaxis for the duration of pregnancy to prevent additional episodes. Other clinicians may reasonably opt to give prophylaxis following two episodes of cystitis, particularly if the time to recurrence is short. Prophylaxis can be postcoital if the cystitis is thought to be sexually related (which it commonly is) or continuous. There are no data to guide whether treatment of isolated recurrent episodes or prophylaxis is a better approach in terms of risks versus benefits.

The choice of antimicrobial used for prophylaxis should be based on the susceptibility profile of the pathogens causing the cystitis. Ideally, daily or postcoital prophylaxis with low-dose nitrofurantoin (50 to 100 mg orally postcoitally or at bedtime) or cephalexin (250 to 500 mg orally postcoitally or at bedtime) can be used.

ACUTE PYELONEPHRITIS

Clinical manifestations — Acute pyelonephritis is a manifestation of infection of the upper urinary tract and kidneys. The typical symptoms of acute pyelonephritis in the pregnant woman are the same as in nonpregnant women and include fever (>38°C or 100.4°F), flank pain, nausea, vomiting, and/or costovertebral angle tenderness. However, focal symptoms of pyelonephritis (eg, flank pain) or of cystitis (eg, dysuria) are not always present. Pyuria is a typical finding.

Most cases of pyelonephritis occur during the second and third trimesters. (See 'Incidence and risk factors' above.)

Pregnant women may become quite ill and are at risk for both medical and obstetrical complications from pyelonephritis. It has been estimated that as many as 20 percent of women with severe pyelonephritis develop complications that include septic shock syndrome or its variants, such as acute respiratory distress syndrome (ARDS) [59-61]. As an example, in a prospective study of 440 cases of acute pyelonephritis identified among 32,282 pregnant women in a general obstetric population, complications included anemia (23 percent), bacteremia (17 percent in the minority of patients who were tested), respiratory insufficiency (7 percent), and renal dysfunction (2 percent) [11]. The mechanism of anemia is not well understood, but hemolysis, perhaps mediated by endotoxin, may be important [62]. Acute renal failure associated with microabscesses and suppurative pyelonephritis has been described in isolated cases, independent of sepsis [63]. (See "Acute kidney injury in pregnancy".)

Diagnosis and evaluation — Acute pyelonephritis is suggested by the presence of flank pain, nausea/vomiting, fever (>38°C or 100.4°F), and/or costovertebral angle tenderness, with or without the typical symptoms of cystitis, and is confirmed by the finding of bacteriuria in the setting of these symptoms. (See "Acute simple cystitis in adult and adolescent females".)

For pregnant women who present with such symptoms, we check a urinalysis and urine culture. Pyuria is present in the majority of women with pyelonephritis, and its absence should prompt consideration of an alternative diagnosis or complete obstruction. However, absence of pyuria does not rule out UTI if symptoms and urine culture are consistent with the diagnosis. Although many pregnant women have back or flank pain without pyelonephritis, we have a low threshold for evaluation for bacteriuria and a diagnosis of pyelonephritis in pregnant women with these symptoms, given the risk of complications and adverse pregnancy outcomes with untreated pyelonephritis. (See 'Pregnancy outcomes' above.)

Some investigators have questioned the value of obtaining routine blood cultures in pregnant women with pyelonephritis [64], and data on the impact of blood cultures on outcomes are limited [65]. Although there is no evidence that bacteremia portends a worse prognosis or requires longer therapy in an otherwise healthy pregnant woman with pyelonephritis, we recommend obtaining blood cultures in those with signs of sepsis or serious underlying medical conditions such as diabetes. Other tests, such as a serum lactate level, can also be useful in women with suspected sepsis to inform the severity of disease [66].

Imaging is not routinely used to diagnose pyelonephritis. However, in patients with pyelonephritis who are severely ill or who also have symptoms of renal colic or history of renal stones, diabetes, history of prior urologic surgery, immunosuppression, repeated episodes of pyelonephritis, or urosepsis, imaging of the kidneys can be helpful to evaluate for complications. In pregnant women, renal ultrasound is the preferred imaging modality in order to avoid contrast or radiation exposure.

Differential diagnosis — Nephrolithiasis can present with significant flank or back pain and abnormal findings on the urinalysis, but fever is uncommon with uncomplicated stone disease. This can also be distinguished from pyelonephritis by visualization of the stones on renal ultrasound. (See "Kidney stones in adults: Diagnosis and acute management of suspected nephrolithiasis".)

For pregnant women presenting with fever and/or flank or back pain, certain obstetric complications are important to consider in the differential:

●Intraamniotic infection, with or without preterm labor, is an important diagnostic consideration in pregnant women who have fever and abdominal pain. The following features suggest intraamniotic infection over pyelonephritis: presentation with premature rupture of membranes, uterine tenderness and/or foul odor of the amniotic fluid, and the absence of bacteriuria. Other potential causes of fever and back or flank pain in the absence of bacteriuria include other infections (eg, influenza, pneumonia, appendicitis). (See "Clinical chorioamnionitis", section on 'Diagnosis'.)

●Placental abruption is a key differential diagnosis of acute back or abdominal pain during pregnancy. Back pain is prominent with abruption when the placenta is on the posterior wall of the uterus. Fever is absent and vaginal bleeding is classically present with abruption, in contrast to pyelonephritis. The uterus is often firm and may be rigid and tender in patients with abruption, but is usually soft in patients with pyelonephritis. Both conditions can be associated with uterine contractions. If present, a retroplacental hematoma on ultrasound supports the diagnosis of abruption. (See "Acute placental abruption: Pathophysiology, clinical features, diagnosis, and consequences".)

Management

Site of care — Based upon the higher risk of complications in pregnant women, pyelonephritis has traditionally been treated with hospitalization and intravenous antibiotics until the woman is afebrile for 24 to 48 hours and symptomatically improved [48]. Initial outpatient therapy of pregnant women with pyelonephritis has been attempted in carefully selected populations (eg, no underlying serious medical conditions, renal or urologic abnormalities, pregnancy complications, signs of sepsis, or recent antibiotic use). However, we suggest not initiating therapy of pyelonephritis in pregnant women in the outpatient setting given the limited data evaluating its safety and the need for close monitoring of the patient.

Evidence on the safety of outpatient initiation of pyelonephritis treatment during pregnancy is limited to two trials by the same group [67,68]. Although the studies suggested that outpatient treatment resulted in similar outcomes as inpatient management, several limitations of the studies create uncertainty about the safety and practicality of outpatient management:

●In the first trial, 120 pregnant women with pyelonephritis prior to 24 weeks gestation were randomly assigned to receive an outpatient regimen consisting of intramuscular ceftriaxone for 2 days followed by oral cephalexin for 10 days or an inpatient regimen consisting of IV cefazolin followed at discharge by oral cephalexin for 10 days [67]. Clinical responses to therapy and birth outcomes were similar for both outpatients and inpatients. However, six patients initially treated with ceftriaxone were ultimately admitted to the hospital for intravenous therapy, and one woman developed septic shock during the emergency department observation period. Of note, the outpatient regimen included initial parenteral antibiotics, and home health nurses monitored patients assigned to the outpatient strategy.

●In the second trial, 92 women who presented after 24 weeks gestation were hospitalized for intramuscular ceftriaxone for 24 hours and then randomly assigned to early discharge on oral cephalexin or continued hospitalization until afebrile for 48 hours [68]. Clinical response and birth outcomes were similar for those who completed the assigned strategy. However, 51 percent of patients either did not qualify for outpatient management based upon study criteria or developed complications, which precluded early discharge from the hospital.

Empiric antibiotics — Parenteral, broad spectrum beta-lactams (eg, ceftriaxone or piperacillin-tazobactam) are the preferred antibiotics for initial empiric therapy of pyelonephritis (table 3). The choice between them should be guided by local microbiology and susceptibility data as well as expected patient tolerance. Fluoroquinolones and aminoglycosides, which are often used for pyelonephritis in nonpregnant individuals, should be avoided in pregnancy if possible. (See 'Antibiotic safety in pregnancy' below.)

The efficacy of beta-lactams was demonstrated in a randomized trial of 179 pregnant women with acute pyelonephritis before the 24th week of gestation: intravenous cefazolin or intramuscular ceftriaxone had equivalent efficacy to intravenous ampicillin plus gentamicin [69]. Although rates of resistance to first generation cephalosporins have generally been less than 10 percent in surveillance studies [70-73], beta-lactams (including first generation cephalosporins) have been less effective than trimethoprim-sulfamethoxazole or the fluoroquinolones for treatment of cystitis in studies of nonpregnant individuals [74]. Given these data and the paucity of data evaluating narrow spectrum cephalosporins in the treatment of pyelonephritis [74], we favor third generation cephalosporins over first or second generation cephalosporins, such as cefazolin, for the empiric treatment of acute pyelonephritis.

For women with a history of infections with extended-spectrum beta-lactamase (ESBL)-producing Enterobacteriaceae (or other risk factors), a carbapenem is an appropriate choice for empiric therapy. Of note, some animal studies have shown adverse fetal effects with imipenem, so meropenem and ertapenem are the preferred carbapenems for use during pregnancy. (See 'Antibiotic safety in pregnancy' below.)

Directed antibiotic therapy and follow-up — As with nonpregnant patients with pyelonephritis, pregnant women generally have definite improvement within 24 to 48 hours of appropriate antibiotic therapy. Once afebrile for 48 hours, pregnant patients can be switched to oral therapy guided by culture susceptibility results and discharged; we typically suggest a total of 7 to 10 days of treatment for patients who have had good clinical response. Data on antibiotic duration for UTI in pregnancy are limited, but evidence in the general population suggest that shorter durations (such as seven days) have comparable clinical outcomes as longer ones. This is discussed elsewhere. (See "Acute complicated urinary tract infection (including pyelonephritis) in adults and adolescents", section on 'Directed antimicrobial therapy'.)

Oral options are mainly limited to beta-lactams or, if in the second trimester, trimethoprim-sulfamethoxazole. Nitrofurantoin and fosfomycin are not appropriate for treatment of pyelonephritis due to inadequate tissue levels. General principles regarding the safety of antibiotics in pregnancy are discussed elsewhere. (See 'Antibiotic safety in pregnancy' below.)

If symptoms and fever persist beyond the first 24 to 48 hours of treatment, a repeat urine culture and renal ultrasound should be performed to rule out persistent infection and urinary tract pathology.

As discussed elsewhere, low-dose antimicrobial prophylaxis following an episode of pyelonephritis is a reasonable option to prevent recurrent pyelonephritis (see 'Preventing recurrence' below). For women who do not use antimicrobial prophylaxis, some clinicians perform monthly urine cultures to evaluate for recurrent bacteriuria and treat as indicated because of the risk of recurrent pyelonephritis. However, the utility of this practice is unclear.

Obstetric management — Pyelonephritis is not itself an indication for delivery. If induction of labor or cesarean delivery for standard obstetrical indications is planned in a patient on treatment for pyelonephritis, we favor waiting until the patient is afebrile, as long as delaying the delivery is relatively safe for the mother and fetus.

Since pyelonephritis is associated with preterm birth, an important obstetric consideration is whether tocolysis should be used when pyelonephritis triggers preterm labor at various gestational ages. Tocolysis is typically not administered after 34 weeks gestation. If a woman with pyelonephritis prior to that gestational age experiences preterm labor, administration of tocolysis and steroids is reasonable to attempt to prolong the pregnancy. However, if the patient is septic, tocolysis is generally avoided. Pregnant women with pyelonephritis are at increased risk of pulmonary edema and acute respiratory distress syndrome (ARDS), which may be exacerbated by administration of tocolysis with or without corticosteroids. Detailed discussion of the benefits and risks of tocolysis for acute preterm labor are found elsewhere. (See "Inhibition of acute preterm labor", section on 'Patient selection'.)

Preventing recurrence — Recurrent pyelonephritis during pregnancy occurs in 6 to 8 percent of women [69,75,76]. As a result, after an initial episode of pyelonephritis, low-dose antimicrobial preventive therapy with an agent to which the original organism is susceptible for the remainder of the pregnancy is a reasonable strategy; but there are no randomized trials to inform the optimal approach. If preventive therapy is utilized, reasonable options include nitrofurantoin (50 to 100 mg orally at bedtime) or cephalexin (250 to 500 mg orally at bedtime) [48,77].

For individuals who use preventive antibiotics following pyelonephritis, we do not perform surveillance urine cultures to ensure effective prophylaxis; data informing this approach are limited.

PREVENTION IN WOMEN WITH HISTORY OF RECURRENT UTI — A separate issue is the management of pregnant women with a history of recurrent urinary tract infections (UTIs) prior to pregnancy, which is often related to sexual intercourse. It is reasonable to use postcoital prophylaxis in pregnant women who have recurrent UTIs that appear to be temporally related to sexual intercourse. The preferred regimen is a single postcoital dose of either cephalexin (250 mg) or nitrofurantoin (50 mg). (See "Recurrent simple cystitis in women", section on 'Initial approach to prevention'.)

The potential efficacy of postcoital prophylaxis to prevent UTIs during pregnancy was evaluated in a report of 33 women with a history of recurrent UTIs who had 39 pregnancies [78]. The women were treated with a single postcoital dose of either cephalexin (250 mg) or nitrofurantoin (50 mg). Only one UTI occurred during pregnancy; this was in sharp contrast to 130 UTIs during a mean observation period of seven months before prophylaxis.

ANTIBIOTIC SAFETY IN PREGNANCY — Penicillins (with or without beta-lactamase inhibitors), cephalosporins, the carbapenems meropenem and ertapenem, aztreonam, and fosfomycin are generally considered safe in pregnancy [79].

Nitrofurantoin is typically avoided during the first trimester because of uncertain associations with congenital anomalies [8,80,81], although overall evidence suggests it is a reasonable alternative during this period of other options cannot be used. The manufacturer indicates that nitrofurantoin is contraindicated at term (eg, 38 to 42 weeks gestation), during labor and delivery, or when labor is imminent because of the possibility of fetal hemolytic anemia [82]. However, data informing this risk are extremely limited, and the American College of Obstetricians and Gynecologists (ACOG) does not restrict nitrofurantoin use during the third trimester [48].

Trimethoprim-sulfamethoxazole is also often avoided during the first trimester because of a potential association with congenital abnormalities and near term because of a theoretical risk of kernicterus [8,80,81]; however, it remains a reasonable alternative during these periods given the uncertainty of the evidence if other options cannot be used.

Aminoglycosides have been associated with ototoxicity following prolonged fetal exposure [83], and fluoroquinolones are generally not used during pregnancy.

Much of the information regarding the safe use of antibiotics during pregnancy was obtained decades ago, before pregnant women were excluded from drug studies because of concerns about risk to the fetus. Thus, there is little direct information about the safety of many newer antibiotics (including many antibiotics that are used for UTI) in pregnancy, and concern about the use of certain antibiotics generally arises from indirect evidence (eg, animal studies) or observational studies that may have numerous confounders. Overall, the safest course is to use the antibiotics that have well-established safety profiles in pregnancy and limit the use of antibiotics of potential concern to cases in which no safer alternative exists.

Additional details on antibiotic safety during pregnancy are found elsewhere. (See "Prenatal care: Patient education, health promotion, and safety of commonly used drugs", section on 'Antibiotics'.)

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Urinary tract infections in adults" and "Society guideline links: Asymptomatic bacteriuria in adults".)

INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5^th to 6^th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10^th to 12^th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.

Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)

●Basics topics (see "Patient education: Urinary tract infections in pregnancy (The Basics)")

●Beyond the Basics topics (see "Patient education: Urinary tract infections in adolescents and adults (Beyond the Basics)")

SUMMARY AND RECOMMENDATIONS

●Risk of bacteriuria – Bacteriuria is common during pregnancy. Without treatment, as many as 30 to 40 percent with asymptomatic bacteriuria develop a symptomatic urinary tract infection (UTI). The smooth muscle relaxation and ureteral dilatation that occurs in pregnancy may facilitate the ascent of bacteria from the bladder to the kidney and increase the risk of pyelonephritis. Additionally, untreated bacteriuria may be associated with an increased risk of preterm birth, low birth weight, and perinatal mortality. (See 'Epidemiology' above and 'Pathogenesis' above.)

●Microbiology – As in nonpregnant women, Escherichia coli is the predominant uropathogen found in both asymptomatic bacteriuria and UTI in pregnant women. (See 'Microbiology' above.)

●Asymptomatic bacteriuria

•Screening – We screen all pregnant women at least once for asymptomatic bacteriuria. Screening for asymptomatic bacteriuria is performed at 12 to 16 weeks gestation with a midstream urine for culture. The diagnosis is made by finding high-level bacterial growth (≥10⁵ colony-forming units [cfu]/mL) on urine culture in the absence of symptoms consistent with UTI. (See 'Diagnosis' above.)

•Management – Management of asymptomatic bacteriuria in pregnant women includes antibiotic therapy tailored to culture results, which reduces the risk of subsequent pyelonephritis and is associated with improved pregnancy outcomes. Potential options include beta-lactams, nitrofurantoin, and fosfomycin (table 1). We do not routinely perform follow-up culture to rescreen for asymptomatic bacteriuria. (See 'Management' above.)

●Cystitis (bladder infection) – Acute cystitis should be suspected in pregnant women who complain about new onset dysuria, frequency, or urgency. The diagnosis is made by finding of bacterial growth on urine culture in this setting. Management of acute cystitis in pregnant women includes empiric antibiotic therapy that is subsequently tailored to culture results. Potential options for empiric and directed therapy include beta-lactams, nitrofurantoin, and fosfomycin (table 1). We do not routinely perform follow-up culture to confirm sterilization of the urine. For those with recurrent cystitis, prophylactic or suppressive antibiotics may be warranted in addition to retreatment. (See 'Acute cystitis' above.)

●Pyelonephritis

•Clinical features and diagnosis – Acute pyelonephritis during pregnancy is suggested by the presence of flank pain, nausea/vomiting, fever (>38°C), and/or costovertebral angle tenderness, with or without the typical symptoms of cystitis, and is confirmed by the finding of bacteriuria in the setting of these symptoms. Pregnant women may become quite ill and are at risk for both medical (eg, sepsis, respiratory failure) and obstetrical complications from pyelonephritis. (See 'Clinical manifestations' above and 'Diagnosis and evaluation' above.)

•Management – Management of acute pyelonephritis in pregnant women includes hospital admission for parenteral antibiotics, preferably broad spectrum beta-lactams (table 3). Antibiotic therapy can be converted to an oral regimen tailored to the susceptibility profile of the isolated organism following clinical improvement. Oral options are generally limited to beta-lactams or, if in the second trimester, trimethoprim-sulfamethoxazole. Following the treatment course, preventive antibiotics are a reasonable strategy for the remainder of the pregnancy to prevent recurrence. (See 'Management' above.)

●Antibiotic safety – It is generally accepted that penicillins (with or without beta-lactamase inhibitors), cephalosporins, aztreonam, and fosfomycin are safe in pregnancy. Because of possible but uncertain associations with adverse birth outcomes, we generally avoid nitrofurantoin during the first trimester and trimethoprim-sulfamethoxazole during the first trimester and near term, but they are appropriate alternatives during those periods if other options cannot be used. (See 'Antibiotic safety in pregnancy' above.)

ACKNOWLEDGMENT — The UpToDate editorial staff acknowledges Thomas M Hooton, MD, who contributed to earlier versions of this topic review.